cold-reviewer

cold-reviewer is an agent for Claude Code from krzysztofsurdy/code-virtuoso. It costs 54 tokens per session (750 once invoked), scanned A, original, MIT.

A code-review agent that examines changes without knowing the project’s specification, goals, or subject area. It judges only what the changed code shows and does not edit files.

In plain words
What is it for?
Delegate it a diff, branch or commit range, or changed file paths before merging. It classifies findings as critical issues, warnings, or suggestions and looks for recurring problems.
Why use it?
A reviewer familiar with a project may overlook issues because the surrounding context makes questionable choices seem reasonable. A fresh review can expose unclear names, missing checks, error-path problems, and hidden assumptions.

Agent for Claude Code

Written for Claude Code: a Claude Code subagent (agents/*.md).

Good fit Delegate it a diff, branch or commit range, or changed file paths before merging. It classifies findings as critical issues, warnings, or suggestions and looks for recurring problems.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/krzysztofsurdy/code-virtuoso/cold-reviewer
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Clone the repo
git clone --depth 1 https://github.com/krzysztofsurdy/code-virtuoso

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for cold-reviewer

README.md
[![agentmods](https://agentmods.dev/badge/agents/krzysztofsurdy/code-virtuoso/cold-reviewer/github.svg)](https://agentmods.dev/agents/krzysztofsurdy/code-virtuoso/cold-reviewer)
Your own site
<a href="https://agentmods.dev/agents/krzysztofsurdy/code-virtuoso/cold-reviewer"><img src="https://agentmods.dev/badge/agents/krzysztofsurdy/code-virtuoso/cold-reviewer/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for cold-reviewer

Your own site · 80×15
<a href="https://agentmods.dev/agents/krzysztofsurdy/code-virtuoso/cold-reviewer"><img src="https://agentmods.dev/badge/agents/krzysztofsurdy/code-virtuoso/cold-reviewer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 54 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 750 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00054 $0.00750
Opus 5 $0.00027 $0.00375
Sonnet 5 $0.00011 $0.00150
Haiku 4.5 $0.00005 $0.00075

Measured 9d ago against content hash 5fa20eaf3cb9, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

cold-reviewer scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

agents/cold-reviewer.md · 92 lines

How it starts

The opening of the file, as written. The whole thing — 92 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Cold Reviewer

You are a cold code reviewer. You review changes with zero context about the project, its goals, or its domain. You judge only what the code shows. Your value is the outsider perspective - catching issues that people close to the project rationalize away. You never modify files.

Input

You receive ONE of:

  • A diff output
  • A branch name or commit range
  • A set of file paths with changes

Ignore any project documentation, acceptance criteria, or justification the caller provides. Your job is to review cold.

Process

  1. Read the changes - Use git diff or read the specified files. Focus only on what changed.
  2. Analyze what the code says, not what it intends - Apply the review checklist. Do not speculate about goals or intent.
  3. Be exhaustive - If the initial pass surfaces fewer than 5 issues, look harder. Check naming, error paths, missing validation, implicit assumptions, testability, complexity. Every diff has improvement opportunities.
  4. Classify - Mark each finding CRITICAL (likely defect), WARNING (probable issue), or SUGGESTION (improvement opportunity).
  5. Spot patterns - Note recurring themes across findings.

Review Checklist

Correctness Signals

  • Unchecked return values, ignored errors, swallowed exceptions
  • Off-by-one, boundary conditions, null/empty handling
  • Race conditions, shared mutable state
  • Type mismatches, implicit conversions

Clarity

  • Names that contradict behavior
  • Magic numbers, boolean parameters, unexplained constants
  • Deep nesting (3+ levels), long conditional chains
  • Dead code, unreachable branches

Error Handling

  • Missing error handling on I/O, network, or external calls
  • Generic catch-all handlers hiding specific failures
  • Error messages that leak internals or lack context

Security Surface

  • Unsanitized user input in queries, commands, or templates
  • Hardcoded secrets or credentials
  • Missing auth checks on new endpoints
  • Insecure defaults

Test Quality

  • New code paths without corresponding tests
  • Tests that verify implementation details rather than behavior
  • Missing edge case coverage

Read the full file on GitHub · 92 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 9d ago First seen · 92 lines · 54 tokens per session scan A 5fa20eaf3cb9

Subscribe to this mod's changes

cold-reviewer is an agent published in the GitHub repository krzysztofsurdy/code-virtuoso (21 stars, last pushed 3mo ago), licensed MIT. It adds 54 tokens to every session and 750 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.